CPOE Studies Using Identical Technology Report Different Results

A new study published in the July issue of Pediatrics finds that the implementation of a computerized physician order entry system at Children’s did not result in an increase in mortality rates.


A new study published in the July issue of Pediatrics finds that the implementation of a computerized physician order entry system at Children’s did not result in an increase in mortality rates.

This study follows a widely publicized study, in the December 2005 issue, that did find an increase in mortality after implementation at Children’s Hospital of Pittsburgh.

Two children’s hospitals’ very different experiences with mortality rates after the implementation of computerized physician order entry systems gives new insight into just how much importance hospitals should place on IT implementation.

CPOE has been touted as an effective way to reduce errors and increase efficiency, so these results should not be surprising. However, this study follows a widely publicized study, published in the December 2005 issue of Pediatrics, that found an increase in mortality after implementation.

The first study reported higher mortality rates among critically ill patients at Children’s Hospital of Pittsburgh after a CPOE system was implemented. The study’s results grabbed headlines and gained national attention.

Pittsburgh implemented Cerner’s Powerchart Orders in 2002 — less than a year before Seattle implemented the same system and reported a “nonstatistically significant” reduction in mortality. So what explains the difference in results?

Dr. Mark Del Beccaro, a Seattle physician and the second study’s lead author, said Seattle decided to conduct the study because it felt that the Pittsburgh study did not provide a complete picture.

He said the hospital considers Pittsburgh a partner and that both hospitals worked very closely together while they were implementing the technology.

Same System, Different Results

At Pittsburgh, the unadjusted patient mortality rate increased from 2.8% before implementation to 6.57% after CPOE implementation during an 18-month period, according to the Pittsburgh study.

At Seattle, the 13-month preimplementation mortality rate was 4.22%, and the 13-month post-implementation mortality rate was 3.46%, according to the Seattle study.

The different outcomes at the two hospitals can be attributed to varying approaches to implementation — including leadership, technical and process factors.

For example, unlike Pittsburgh, Seattle “had active involvement of [the] intensive care unit staff during the design, build and implementation stages,” according to the study.

implementation issues…rather than inherent issues with the CPOE itself…are the primary risk factors affecting mortality

Also, “Both institutions placed a great deal of effort in designing and implementing order sets, but CHP did not have the order sets for the critical care setting available at implementation,” the CHRMC study notes.

According to the Seattle study, “implementation issues…rather than inherent issues with the CPOE itself…are the primary risk factors affecting mortality during implementation of CPOE.

Del Beccaro notes that CHP did not have the benefit of extensive previous data or studies to use as a model, so “some of the things they learned were by trial and error.”

Lessons Learned

Together, the two papers show that “these are the things you should do, and these are some of the things you shouldn’t do,” Del Beccaro said.

For example, the study notes that Seattle was concerned with CPOE’s potential to cause a breakdown in communication and that the hospital has a mantra, “CPOE does not replace talking.”

At Seattle, Del Beccaro said the CPOE system has:

  • Eliminated handwriting errors
  • Accelerated medication turnaround time
  • Enhanced the hospital’s process improvement initiative
  • Helped standardize care

A key to Seattle’s success was that the hospital made CPOE implementation a culture change for the entire organization, according to Del Beccaro.

Everyone was involved in the process, and the entire hospital went live with the system at the same time, he said. “One of the benefits for that is that you actually do grab the entire institution’s attention and direction,” Del Beccaro said.

Del Beccaro said it is important for hospitals to make their commitment to the technology obvious to end users, stressing that it is not just a pilot program. He also said it is crucial to get everyone involved. “If you don’t do it as an institutional culture change, your risk of failure goes way, way up.”

The study concludes that Seattle’s “experience suggests that careful design, build, implementation and support can mitigate the risk of implementing new technology, even in an ICU setting.”

About Seattle Children’s

Consistently ranked as one of the best children’s hospitals in the country by U.S. News & World Report, Seattle Children’s serves as the pediatric and adolescent academic medical referral center for the largest landmass of any children’s hospital in the country (Washington, Alaska, Montana and Idaho). For more than 100 years, Seattle Children’s has been delivering superior patient care while advancing new treatments through pediatric research. Seattle Children’s serves as the primary teaching, clinical and research site for the Department of Pediatrics at the University of Washington School of Medicine. The hospital works in partnership with Seattle Children’s Research Institute and Seattle Children’s Hospital Foundation. For more information, visit www.seattlechildrens.org or follow us on Twitter or Facebook.